Main Engine Cut Off

Atlas V Dual Engine Centaur Performance

In the wake of the announcement that Dream Chaser will fly to the ISS on an Atlas V 552, I’ve been curious how much performance the dual engine Centaur will add to the Atlas V configurations we know and love.

I bugged Tory Bruno about it and he delivered:

@WeHaveMECO @ethan829 An extra RL10 would add another 4000lbm (1800 kg) of payload mass

Pretty impressive performance improvement. Doubling the low thrust of a single RL10 reduces gravity losses quite a bit and adds almost two metric tons to lift capacity.

So far, there are only 3 payloads that will use the dual engine Centaur. Boeing’s Starliner will fly an N22 configuration of the Atlas V, which means no fairing, 2 solid boosters, and 2 RL10 engines. Dream Chaser and Bigelow’s B330 will fly a 552 configuration—a 5 meter fairing, 5 solid boosters, and 2 RL10 engines.

I’m hoping to see the dual engine Centaur added to ULA’s RocketBuilder so we can get an idea of how much more an extra RL10 costs. That may be our first real look at just how much that historic engine costs these days.

T+53: Loren Grush

Farewell, Red Dragon

A few weeks ago, I caught wind1 of the news: Dragon 2 propulsive landings had been cancelled2 for both cargo and crew, and thus, Red Dragon was over. The margins were already pretty tight for landings—tight enough to make NASA uncomfortable—and the integrated landing leg-heat shield assembly was problematic.

After a few weeks spent digesting the decision, I can understand the reasoning. Designing and developing solutions to the existing problems would take a lot of time and money. The SpaceX style is to develop their systems iteratively as part of the missions they fly for customers—most notably, the reusability of Falcon 9’s first stage which was developed, tested, and implemented as they flew missions for NASA and others.

Without the ability to develop, test, and implement Dragon 2 propulsive landings on NASA flights, their budget and timeline expanded into unacceptable territory. The NASA decision alone may have added two or more additional years, pushing the first flight to 2022 or beyond, and several hundred million dollars—to cover the cost of engineers, hardware, test flights, and so on—to the Red Dragon project. And needless to say, that would have pushed off the first flight of the Interplanetary Transport System quite a bit.

Red Dragon was a great project as a spinoff of capability developed as part of SpaceX’s work for NASA, but it could not justify the relatively massive additional development cost and time on its own, because it was a developmental dead end. There is almost zero commonality between Red Dragon and ITS—the two spacecraft have entirely different body types, entry interfaces, atmospheric flight profiles, engine types, engine arrangement, control systems, materials, and on, and on.

SpaceX’s Mars-focused resources would be better spent on ITS, and they’re going to double down on it. They’re changing the architecture to accelerate the switch to their next-generation system. My guess is that their target is to have the Raptor-based, fully reusable system flying by 2021 at the earliest, with the first uncrewed mission to Mars in 2022 or 2024.

With that, Red Dragon joins the pile of projects SpaceX has scrapped in pursuit of their vision—Falcon 1, Falcon 1e, Falcon 5, and Falcon 9’s parachutes, to name the infamous few. When circumstances change and they find that something much better is just inches from their fingertips, SpaceX is quick, decisive, and goes all in on the new path. Progress is messy.

And while I understand the reasoning behind this decision, and while I’m excited to see the new path SpaceX has chosen, and while I’m optimistic that the new path will be even better, I can’t help but be hugely disappointed.

The first successful landing of Red Dragon would have been a watershed moment in the history of space. It would have changed everything to see a human-scale spacecraft land on Mars for the first time. More so to have it be a spacecraft that is actively being used to fly astronauts to the ISS. And even more so to have that spacecraft be developed and operated by a private company. It would have changed everything.

For now, we wait. Maybe we wait another 3 or 4 transfer windows.

But that watershed moment is coming. It’s inevitable.

  1. I don’t think they wanted to talk about this until they unveiled the new architecture, but it had started to leak publicly a few days before Elon’s talk. Once it started leaking, SpaceX seems to have thought it better to get out in front of it and be able to shape the story, rather than having rumors percolate for months and have expectations build. ↩︎

  2. This decision also explains why we hadn’t seen a hover test since November, 2015, and why Gwynne Shotwell was vague about Red Dragon on The Space Show. ↩︎

Sierra Nevada To Fly on Atlas V 552

The two awarded Atlas V missions will carry pressurized and unpressurized cargo to the International Space Station (ISS). The first mission is set to lift off in 2020 from Space Launch Complex 41 at Cape Canaveral Air Force Station, in Florida. The second contracted mission is scheduled to lift off in 2021. Dream Chaser will launch atop an Atlas V 552, with a dual engine Centaur upper stage.

A launch on an Atlas V 552 probably runs pretty damn close to the $200 million mark.


The Reusability Inflection Point

Each milestone achieved in the march towards launch vehicle reusability brought with it a slew of “This is the moment!” articles. The technical achievements—the first successful landing of Falcon 9, the first relaunch of Falcon 9, and so on—have all been huge moments, to be sure.

But, at the risk of sounding like an Ariane executive, the true inflection point is an economic one. Not an arbitrary economic milestone like making launch costs half of what they are today—it’s about the business model of reusability and how it’s integrated into a company’s operations.


The current pricing SpaceX uses makes no sense in the high-level, long-term view. It’s pretty obvious that the model grew out of an expendable-minded era, and that SpaceX is sticking to it because the market itself has not yet changed its thinking.

If you go to buy a launch on a Falcon 9 today, you have two choices: $60 million(ish) for a flight with a new first stage, or $50 million(ish) for a flight with a previously-flown first stage.

The model should look like this: $60 million for a flight in which SpaceX cannot recover the first stage, and $50 million for a flight in which SpaceX can recover the first stage. And to push the model even further, a flight with a possible return-to-launch-site recovery should cost less than a flight which requires the droneship.

To use Elon’s favorite ill-fitting, horribly-tortured example of a plane flight: your ticket price should be based on your destination, fuel usage, and cargo requirements rather than based on where that plane’s previous flight was headed.

In this new model, a customer wouldn't know and, more importantly, shouldn’t care whether the first stage that flies their payload is making its first or fifth flight. That decision becomes part of SpaceX’s internal fleet management operations, the same way that hardware and procedural upgrades are handled today. Iridium-2 flew with new titanium grid fins—admittedly a nearly-irrelevant piece of hardware during ascent—and I’m sure that wasn’t commented on within their launch contract.

In order to implement and embrace this new model, at least two main areas need to be addressed.

Reliability and Operations

It’s not a shocking statement to say that the reliability of Falcon 9 needs to improve quite a bit overall for any of these economic considerations to matter. Two lost vehicles and payloads in two years is horrible, and SpaceX needs to avoid anything like that happening for quite some time to build trust.

Moreover, SpaceX needs to prove that reused stages are just as—if not more—reliable than a brand new stage. That’s the single biggest mental block the market at large needs to get over for the era of reusability to really come into its own.

And while SpaceX has gotten far more stages back than they had predicted, they’re not without loss. Given the sea conditions they land in from time to time, that’s to be expected.

For the new model to be successful, SpaceX needs to get their reliability of recovery to a point where they lose as few stages as possible, but they also need to get their operations to a point where they can absorb the occasional loss. They need to be getting enough flights out of each stage to amortize its production and eventually turn a profit from it, so much so that a recovery loss of a first-flight stage doesn’t set them back drastically.


With all that said, when should SpaceX make the pricing change? Surely 2017 is too soon—the final version of Falcon 9 isn’t yet flying, the recovery operations are still being perfected, their additional inspection, refurbishment, and storage facilities at the Cape aren’t yet ready, and the government is still figuring out how to deal with flown stages.

But at some point in the very near future, SpaceX will have to decide that enough of the above to-do list is complete, that their recovery rate is high enough, and that they’ve proven the reliability of previously-flown stages.

It’ll be a big, bold moment. It’ll probably scare off some customers, and almost everyone will be saying that it’s far too soon to shift the reusability mindset.

But it’s exactly what SpaceX needs to do.

Starliner Free Flight (In)Capability

Chris Gebhardt of had a great piece on Starliner’s current status:

The latest confirmed schedules from NASA show the uncrewed mission, dubbed the Orbital Flight Test (OFT), slated for No Earlier Than June 2018, followed quickly in August 2018 by the crewed flight test.

However, comments made by Chris Ferguson last month at the Paris Air Show seem to indicate that the crewed flight test is moving from its August timeframe.

According to Mr. Ferguson, Director of Crew and Mission Operations for Boeing’s Commercial Crew Program, the first Starliner crewed test flight is aiming for “last quarter of 2018” – which would be a shift of two to five months into the October to December 2018 timeframe.

Later on in the article:

Here, Mr. Ferguson noted his desire for Starliner to perform the rapid sequence 6 hour launch to docking profile currently employed by Soyuz crew vehicles and Progress uncrewed resupply vehicles from Russia.

At most, Mr. Ferguson stated his desire for Starliner to employ 24-hour launch to docking profiles – due in part to the vehicle’s design, which limits its free flight capability (from launch to docking and then undocking to landing) for an entire mission to just 60 hours.

It seems that Boeing took Charlie Bolden too literally: his oft-heard refrain was that NASA turned over LEO to commercial companies and wanted beyond LEO to itself. Boeing designed and built a vehicle that literally can not fly beyond LEO without help.

SpaceX, on the other hand, built a vehicle with much more capability than was explicitly asked for by NASA—they’re working towards a private lunar free-return flight with Dragon 2 that will most likely piss off certain people within NASA if they do it before Orion can get there with NASA astronauts aboard.

SpaceX took the opportunity to build capability that would be useful—and lucrative—beyond their ISS contract. After all, that is the point of any program aimed to be commercial by name. Boeing built just what was needed and nothing more, and that’s going to haunt them if and when they pursue anything other than NASA crew contracts.

Between the weak free flight capability and the expensive launch vehicle—an Atlas V 422, which means paying for two solid boosters and two RL10 engines—I’m forced to ask again questions I asked long ago: what the hell else is Starliner going to do?

More Details on Sierra Nevada’s Dream Chaser and Deep Space Gateway Concept

The ISS R&D Conference is kicking off this week in DC with sure-to-be interesting keynotes by Elon Musk, Robert Bigelow, and a few members of Congress, among others. But there are also a bunch of technical sessions in the afternoons, so I went digging through the agenda. I found two sessions by Sierra Nevada—one on Dream Chaser at the ISS and one on their NextSTEP-2 Deep Space Gateway concept. Lucky for us, the session PDFs are up already.

The session focused on science opportunities using Dream Chaser at ISS (PDF, 1.3MB) has a few things of note—either new information, more details on things we’ve heard in the past, or things I had forgotten until now:

  • Dream Chaser has 10+ day on-orbit free flight time. That’s a decent chunk of time, unlike the 60-hour limit for Starliner.
  • Its propulsion system can be used for ISS reboost.
  • The cargo module has 3 FRAMs for smaller, unpressurized cargo.
  • They’re advertising the cargo module as a temporary lab, too: ”Cargo module could be modified and left at station as desired. It could be launched with additional specialized facilities: Centrifuge, Rodent habitats, Glove boxes, etc.”
  • They also mention several service kits that can be added to the cargo module for additional capabilities:
    • Modified solar arrays
    • Dedicated battery packs
    • Deployable radiators
    • Upgraded antennas
    • Additional data storage
    • Dedicated payload downlinks
    • Optical observation windows
    • Unpressurized payload return

The session focused on testing their NextSTEP-2 Deep Space Gateway concept at the ISS (PDF, 1MB) is worth a look too see the basic layout of their concept. It’s based around Dream Chaser’s cargo module with an inflatable airlock and an inflatable habitat. They also mention and show a commercial LEO station comprised of 4 cargo modules with the potential addition of inflatable habitats.

Two thoughts that went through my mind as I went through the PDF:

  • Their gateway concept is really disjointed looking.
  • It’s interesting to see someone other than Bigelow (and NASA, about Bigelow) talking about inflatables.

T+52: XCOR Winds Down, ULA Wins STP-3

XCOR laid off the rest of its staff and is closing up shop after losing a contract with ULA, which leaves ULA in an interesting spot for Vulcan-ACES. On the ULA side, they won their first Phase 1A contract from the Air Force, and the contract price sheds some light on just how much they’re cutting their costs.

This episode of Main Engine Cut Off is brought to you by 16 executive producers—Kris, Mike, Pat, Matt, Jorge, Brad, Ryan, Laszlo, Jamison, Guinevere, Nadim, Peter, and four anonymous—and 56 other supporters on Patreon.

Scottish Spaceport Proposed

Russell Jackson, for The Scotsman:

A consortium, which includes US aerospace firm Lockheed Martin, believe that the A’Mhoine peninsula in Sutherland would be the ideal location in Britain from which satellites could be launched into orbit.

A detailed proposal for the facility - located between Dounreay and Cape Wrath - has been submitted to the UK Space Agency (UKSA) which has met with Highland council and Highlands and Islands Entreprise (HIE) to discuss the plans.

The Scottish base - which could be operational by 2020 - would be the first to launch a rocket into space from UK soil.

I always thought the north side of the UK would be a great spot for a polar launch site.

Some will liken this to the spaceport dilemma in the US, but it’s totally different. The troubled spaceports in the US all have a critical flaw in them for frequent commercial use: Spaceport America isn’t able to support orbital launches from its location (Mojave, too, for that matter), the Pacific Spaceport Complex is too remote, Wallops can only support highly-inclined-but-not-polar launches, and so on. Georgia may find itself in a similar situation as Wallops—unable to support polar launches and worse for GEO launches than Canaveral.

A polar orbit-specific launch site on UK soil has real potential.

JAXA Interested in NASA’s Deep Space Gateway

Seiji Tanaka, for The Asahi Shimbun:

The Japan Aerospace Exploration Agency (JAXA) revealed its ambitious plan for a Japanese astronaut to set foot on the moon around 2030.

But rather than developing its own manned space rocket, and shoot to the moon, JAXA plans on a much shorter trip.

It hopes that a U.S. space station planned to go into lunar orbit will allow Japanese astronauts to hop off and on to the moon around 2030.

This is exactly what NASA was hoping to achieve by putting the plans for the Deep Space Gateway out into the public eye. They need international and commercial partners to latch onto the idea of the Deep Space Gateway, develop plans to use it, and talk about those plans publicly. That’s how they can build support for it within the US political sphere, and that’s how they can get it funded.

Lunar surface access is exactly the gap they were hoping for partners to step up and fill in, and there’s already interest from Blue Origin and now JAXA.